Kisitli ses külliyati ile Türkçe geniş daǧarcikli sürekli konuşma tanima

Derya Susman; Selçuk Köprü; Adnan Yazici

doi:10.1109/SIU.2012.6204601

Kisitli ses külliyati ile Türkçe geniş daǧarcikli sürekli konuşma tanima

Translated title of the contribution: Turkish large vocabulary continuous speech recognition by using limited audio corpus

Derya Susman, Selçuk Köprü, Adnan Yazici

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Citations (Scopus)

Abstract

In this paper, the recognition performances of several methodologies proposed in the context of Turkish Large Vocabulary Continuous Speech Recognition are retrieved by using a limited audio corpus. Word based, stem based, stem-ending based, and morph based language models are utilized with different n-gram orders. Word based and stem-ending based language models are extended by using several approaches. Also, a hybrid language model which is based on word based and stem-ending based language models is proposed. Word based language model is observed to outperform sub-word language models when limited audio corpus is used.

Translated title of the contribution	Turkish large vocabulary continuous speech recognition by using limited audio corpus
Original language	Undefined/Unknown
Title of host publication	2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings
DOIs	https://doi.org/10.1109/SIU.2012.6204601
Publication status	Published - 2012
Externally published	Yes
Event	2012 20th Signal Processing and Communications Applications Conference, SIU 2012 - Fethiye, Mugla, Turkey Duration: Apr 18 2012 → Apr 20 2012

Publication series

Name	2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings

Conference

Conference	2012 20th Signal Processing and Communications Applications Conference, SIU 2012
Country/Territory	Turkey
City	Fethiye, Mugla
Period	4/18/12 → 4/20/12

Keywords

agglutinative
hidden markov model
large vocabulary continuous speech recognition
limited corpus
n-gram language model

ASJC Scopus subject areas

Computer Networks and Communications
Signal Processing

Access to Document

10.1109/SIU.2012.6204601

Cite this

Susman, D., Köprü, S., & Yazici, A. (2012). Kisitli ses külliyati ile Türkçe geniş daǧarcikli sürekli konuşma tanima. In 2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings Article 6204601 (2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings). https://doi.org/10.1109/SIU.2012.6204601

Kisitli ses külliyati ile Türkçe geniş daǧarcikli sürekli konuşma tanima. / Susman, Derya; Köprü, Selçuk; Yazici, Adnan.
2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings. 2012. 6204601 (2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Susman, D, Köprü, S & Yazici, A 2012, Kisitli ses külliyati ile Türkçe geniş daǧarcikli sürekli konuşma tanima. in 2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings., 6204601, 2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings, 2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Fethiye, Mugla, Turkey, 4/18/12. https://doi.org/10.1109/SIU.2012.6204601

@inproceedings{c9f3ccff79444c109ef60c8797357315,

title = "Kisitli ses k{\"u}lliyati ile T{\"u}rk{\c c}e geni{\c s} daǧarcikli s{\"u}rekli konu{\c s}ma tanima",

abstract = "In this paper, the recognition performances of several methodologies proposed in the context of Turkish Large Vocabulary Continuous Speech Recognition are retrieved by using a limited audio corpus. Word based, stem based, stem-ending based, and morph based language models are utilized with different n-gram orders. Word based and stem-ending based language models are extended by using several approaches. Also, a hybrid language model which is based on word based and stem-ending based language models is proposed. Word based language model is observed to outperform sub-word language models when limited audio corpus is used.",

keywords = "agglutinative, hidden markov model, large vocabulary continuous speech recognition, limited corpus, n-gram language model",

author = "Derya Susman and Sel{\c c}uk K{\"o}pr{\"u} and Adnan Yazici",

year = "2012",

doi = "10.1109/SIU.2012.6204601",

language = "Undefined/Unknown",

isbn = "9781467300568",

series = "2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings",

booktitle = "2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings",

note = "2012 20th Signal Processing and Communications Applications Conference, SIU 2012 ; Conference date: 18-04-2012 Through 20-04-2012",

}

TY - GEN

T1 - Kisitli ses külliyati ile Türkçe geniş daǧarcikli sürekli konuşma tanima

AU - Susman, Derya

AU - Köprü, Selçuk

AU - Yazici, Adnan

PY - 2012

Y1 - 2012

N2 - In this paper, the recognition performances of several methodologies proposed in the context of Turkish Large Vocabulary Continuous Speech Recognition are retrieved by using a limited audio corpus. Word based, stem based, stem-ending based, and morph based language models are utilized with different n-gram orders. Word based and stem-ending based language models are extended by using several approaches. Also, a hybrid language model which is based on word based and stem-ending based language models is proposed. Word based language model is observed to outperform sub-word language models when limited audio corpus is used.

AB - In this paper, the recognition performances of several methodologies proposed in the context of Turkish Large Vocabulary Continuous Speech Recognition are retrieved by using a limited audio corpus. Word based, stem based, stem-ending based, and morph based language models are utilized with different n-gram orders. Word based and stem-ending based language models are extended by using several approaches. Also, a hybrid language model which is based on word based and stem-ending based language models is proposed. Word based language model is observed to outperform sub-word language models when limited audio corpus is used.

KW - agglutinative

KW - hidden markov model

KW - large vocabulary continuous speech recognition

KW - limited corpus

KW - n-gram language model

UR - http://www.scopus.com/inward/record.url?scp=84863501240&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863501240&partnerID=8YFLogxK

U2 - 10.1109/SIU.2012.6204601

DO - 10.1109/SIU.2012.6204601

M3 - Conference contribution

AN - SCOPUS:84863501240

SN - 9781467300568

T3 - 2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings

BT - 2012 20th Signal Processing and Communications Applications Conference, SIU 2012, Proceedings

T2 - 2012 20th Signal Processing and Communications Applications Conference, SIU 2012

Y2 - 18 April 2012 through 20 April 2012

ER -

Kisitli ses külliyati ile Türkçe geniş daǧarcikli sürekli konuşma tanima

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this